文章目錄
- 一、安裝dlib庫
- 二、利用dlib實作人臉68個關鍵點檢測并标注
- 三、人臉特征提取
- 四、人臉識别
- 參考連結
環境說明：
python3.6+spyder
第三方庫的說明
skimage，playsound
安裝指令：
pip install -i https://pypi.tuna.tsinghua.edu.cn/simple scikit-image
pip install -i https://pypi.tuna.tsinghua.edu.cn/simple playsound
一、安裝dlib庫
-
下載下傳Dlib安裝包
下載下傳連結：http://dlib.net/files/
本文章下載下傳的是
dlib-19.14.zip
-
安裝Cmake
下載下傳連結：https://cmake.org/download/
下載下傳安裝包直接點選安裝就行，注意環境變量的設定
-
下載下傳boost
下載下傳連結：http://www.boost.org/
下載下傳之後将其解壓縮，進入解壓後的檔案夾中，找到bootstrap.bat批處理檔案，輕按兩下運作，等待運作完成後（指令行自動消失）會生成檔案b2.exe
win+R，打開指令行，進入b2.exe所在的檔案夾，運作下面指令 b2編譯庫檔案
安裝完成後配置boost環境變量b2 -a -python address-model=64 toolset=msvc runtime-link=static #cmake下載下傳的64位這裡（address-model）寫64，如果是32位的就把之前的64改成32
Dlib模型實作人臉識别一、安裝dlib庫二、利用dlib實作人臉68個關鍵點檢測并标注三、人臉特征提取四、人臉識别參考連結 -
安裝dlib
指令行進入dlib的檔案夾中 安裝完成後，在檔案夾下面會出現dlib，dlib.egg-info，dist的三個檔案夾
将dlib 和dlib.egg-info 複制對應python環境下的Lib檔案，同時将build\lib.win-amd64-3.6檔案夾下的dlib.cp36-win_amd64.pyd複制到對應python環境下的DLL檔案夾
測試是否安裝成功（沒有報錯，表示安裝成功）
Dlib模型實作人臉識别一、安裝dlib庫二、利用dlib實作人臉68個關鍵點檢測并标注三、人臉特征提取四、人臉識别參考連結
二、利用dlib實作人臉68個關鍵點檢測并标注
-
下載下傳官方的訓練模型
下載下傳連結：
http://dlib.net/files/shape_predictor_68_face_landmarks.dat.bz2
- 人臉檢測和标注
标注結果import numpy as np import cv2 import dlib #detector = dlib.get_frontal_face_detector() detector=dlib.get_frontal_face_detector() predictor = dlib.shape_predictor('E:\\PersonRecognitionDlib\\shape_predictor_68_face_landmarks.dat\\shape_predictor_68_face_landmarks.dat') # cv2讀取圖像 img = cv2.imread("E:\\PersonRecognitionDlib\\text.jpg") #print(img) # 取灰階 img_gray = cv2.cvtColor(img, cv2.COLOR_RGB2GRAY) # 人臉數rects rects = detector(img_gray, 1) for i in range(len(rects)): landmarks = np.matrix([[p.x, p.y] for p in predictor(img,rects[i]).parts()]) for idx, point in enumerate(landmarks): # 68點的坐标 pos = (point[0, 0], point[0, 1]) print(idx,pos) # 利用cv2.circle給每個特征點畫一個圈，共68個 cv2.circle(img, pos, 5, color=(0, 255, 0)) # 利用cv2.putText輸出1-68 font = cv2.FONT_HERSHEY_SIMPLEX cv2.putText(img, str(idx+1), pos, font, 0.8, (0, 0, 255), 1,cv2.LINE_AA) cv2.namedWindow("img", 2) cv2.imshow("img", img) cv2.waitKey(0)問題描述： Dlib模型實作人臉識别一、安裝dlib庫二、利用dlib實作人臉68個關鍵點檢測并标注三、人臉特征提取四、人臉識别參考連結 module 'dlib' has no attribute 'get_frontal_face_detector'
三、人臉特征提取
-
人臉資料集
①使用攝像頭采集（視訊流截圖）
采集的過程，最好使用同一裝置同一光線下進行采集
import cv2 import dlib import os import sys import random # 存儲位置 output_dir = 'D:/myworkspace/JupyterNotebook/People/person/person1' size = 64 if not os.path.exists(output_dir): os.makedirs(output_dir) # 改變圖檔的亮度與對比度 def relight(img, light=1, bias=0): w = img.shape[1] h = img.shape[0] #image = [] for i in range(0,w): for j in range(0,h): for c in range(3): tmp = int(img[j,i,c]*light + bias) if tmp > 255: tmp = 255 elif tmp < 0: tmp = 0 img[j,i,c] = tmp return img #使用dlib自帶的frontal_face_detector作為我們的特征提取器 detector = dlib.get_frontal_face_detector() # 打開攝像頭 參數為輸入流，可以為攝像頭或視訊檔案 camera = cv2.VideoCapture(0) index = 1 while True: if (index <= 15):#存儲15張人臉特征圖像 print('Being processed picture %s' % index) # 從攝像頭讀取照片 success, img = camera.read() # 轉為灰階圖檔 gray_img = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY) # 使用detector進行人臉檢測 dets = detector(gray_img, 1) for i, d in enumerate(dets): x1 = d.top() if d.top() > 0 else 0 y1 = d.bottom() if d.bottom() > 0 else 0 x2 = d.left() if d.left() > 0 else 0 y2 = d.right() if d.right() > 0 else 0 face = img[x1:y1,x2:y2] # 調整圖檔的對比度與亮度， 對比度與亮度值都取随機數，這樣能增加樣本的多樣性 face = relight(face, random.uniform(0.5, 1.5), random.randint(-50, 50)) face = cv2.resize(face, (size,size)) cv2.imshow('image', face) cv2.imwrite(output_dir+'/'+str(index)+'.jpg', face) index += 1 key = cv2.waitKey(30) & 0xff if key == 27: break else: print('Finished!') # 釋放攝像頭 release camera camera.release() # 删除建立的視窗 delete all the windows cv2.destroyAllWindows() break Dlib模型實作人臉識别一、安裝dlib庫二、利用dlib實作人臉68個關鍵點檢測并标注三、人臉特征提取四、人臉識别參考連結 在對應的輸出目錄下，會得到15張攝像頭采集得到的圖檔。
②網絡爬蟲擷取
具體内容可以參考連結：
https://blog.csdn.net/cungudafa/article/details/87862687
- 資料集的處理
-
擷取特征點
①下載下傳dlib的人臉識别模型
下載下傳連結：
https://pan.baidu.com/s/1sBH4TvIfIYLFYs7zCTH4nA
提取碼：b8zu
②擷取每個人68個特征資料并儲存到csv中
from cv2 import cv2 as cv2 import os import dlib from skimage import io import csv import numpy as np # 要讀取人臉圖像檔案的路徑 path_images_from_camera = "E:/PersonRecognitionDlib/Person/" # Dlib 正向人臉檢測器 detector = dlib.get_frontal_face_detector() # Dlib 人臉預測器 predictor = dlib.shape_predictor("E:/PersonRecognitionDlib/model/shape_predictor_68_face_landmarks.dat") # Dlib 人臉識别模型 # Face recognition model, the object maps human faces into 128D vectors face_rec = dlib.face_recognition_model_v1("E:/PersonRecognitionDlib/model/dlib_face_recognition_resnet_model_v1.dat") # 傳回單張圖像的 128D 特征 def return_128d_features(path_img): img_rd = io.imread(path_img) img_gray = cv2.cvtColor(img_rd, cv2.COLOR_BGR2RGB) faces = detector(img_gray, 1) print("%-40s %-20s" % ("檢測到人臉的圖像 / image with faces detected:", path_img), '\n') # 因為有可能截下來的人臉再去檢測，檢測不出來人臉了 # 是以要確定是 檢測到人臉的人臉圖像 拿去算特征 if len(faces) != 0: shape = predictor(img_gray, faces[0]) face_descriptor = face_rec.compute_face_descriptor(img_gray, shape) else: face_descriptor = 0 print("no face") return face_descriptor # 将檔案夾中照片特征提取出來, 寫入 CSV def return_features_mean_personX(path_faces_personX): features_list_personX = [] photos_list = os.listdir(path_faces_personX) if photos_list: for i in range(len(photos_list)): # 調用return_128d_features()得到128d特征 print("%-40s %-20s" % ("正在讀的人臉圖像 / image to read:", path_faces_personX + "/" + photos_list[i])) features_128d = return_128d_features(path_faces_personX + "/" + photos_list[i]) # print(features_128d) # 遇到沒有檢測出人臉的圖檔跳過 if features_128d == 0: i += 1 else: features_list_personX.append(features_128d) else: print("檔案夾内圖像檔案為空 / Warning: No images in " + path_faces_personX + '/', '\n') # 計算 128D 特征的均值 # N x 128D -> 1 x 128D if features_list_personX: features_mean_personX = np.array(features_list_personX).mean(axis=0) else: features_mean_personX = '0' return features_mean_personX # 讀取某人所有的人臉圖像的資料 people = os.listdir(path_images_from_camera) people.sort() with open("E:/PersonRecognitionDlib/feature/features2_all.csv", "w", newline="") as csvfile: writer = csv.writer(csvfile) for person in people: print("##### " + person + " #####") # Get the mean/average features of face/personX, it will be a list with a length of 128D features_mean_personX = return_features_mean_personX(path_images_from_camera + person) writer.writerow(features_mean_personX) print("特征均值 / The mean of features:", list(features_mean_personX)) print('\n') print("所有錄入人臉資料存入 / Save all the features of faces registered into: D:/myworkspace/JupyterNotebook/People/feature/features_all2.csv") Dlib模型實作人臉識别一、安裝dlib庫二、利用dlib實作人臉68個關鍵點檢測并标注三、人臉特征提取四、人臉識别參考連結
四、人臉識别
-
計算距離——歐氏距離
将要識别的對象與已經存在的對象進行距離計算
def return_euclidean_distance(feature_1, feature_2): feature_1 = np.array(feature_1) feature_2 = np.array(feature_2) dist = np.sqrt(np.sum(np.square(feature_1 - feature_2))) return dist - 實作人臉識别
整個過程是先進行人臉的檢測，檢測攝像頭所采集到的人臉，然後将檢測到的人臉對象與資料集中的人臉68個特征點進行一個距離計算，然後，選出最接近的那個人，接着判斷距離小于0.4，就可以辨別出識别人物。# 攝像頭實時人臉識别 import os import winsound # 系統音效 from playsound import playsound # 音頻播放 import dlib # 人臉處理的庫 Dlib import csv # 存入表格 import time import sys import numpy as np # 資料處理的庫 numpy from cv2 import cv2 as cv2 # 圖像處理的庫 OpenCv import pandas as pd # 資料處理的庫 Pandas # 人臉識别模型，提取128D的特征矢量 # face recognition model, the object maps human faces into 128D vectors # Refer this tutorial: http://dlib.net/python/index.html#dlib.face_recognition_model_v1 facerec = dlib.face_recognition_model_v1("E:/PersonRecognitionDlib/model/dlib_face_recognition_resnet_model_v1.dat") # 計算兩個128D向量間的歐式距離 # compute the e-distance between two 128D features def return_euclidean_distance(feature_1, feature_2): feature_1 = np.array(feature_1) feature_2 = np.array(feature_2) dist = np.sqrt(np.sum(np.square(feature_1 - feature_2))) return dist # 處理存放所有人臉特征的 csv path_features_known_csv = "E:/PersonRecognitionDlib/feature/features2_all.csv" csv_rd = pd.read_csv(path_features_known_csv, header=None) # 用來存放所有錄入人臉特征的數組 # the array to save the features of faces in the database features_known_arr = [] # 讀取已知人臉資料 # print known faces for i in range(csv_rd.shape[0]): features_someone_arr = [] for j in range(0, len(csv_rd.iloc[i, :])): features_someone_arr.append(csv_rd.iloc[i, :][j]) features_known_arr.append(features_someone_arr) print("Faces in Database：", len(features_known_arr)) # Dlib 檢測器和預測器 # The detector and predictor will be used detector = dlib.get_frontal_face_detector() predictor = dlib.shape_predictor('E:/PersonRecognitionDlib/model/shape_predictor_68_face_landmarks.dat') # 建立 cv2 攝像頭對象 # cv2.VideoCapture(0) to use the default camera of PC, # and you can use local video name by use cv2.VideoCapture(filename) cap = cv2.VideoCapture(0) # cap.set(propId, value) # 設定視訊參數，propId 設定的視訊參數，value 設定的參數值 cap.set(3, 480) # cap.isOpened() 傳回 true/false 檢查初始化是否成功 # when the camera is open while cap.isOpened(): flag, img_rd = cap.read() kk = cv2.waitKey(1) # 取灰階 img_gray = cv2.cvtColor(img_rd, cv2.COLOR_RGB2GRAY) # 人臉數 faces faces = detector(img_gray, 0) # 待會要寫的字型 font to write later font = cv2.FONT_HERSHEY_COMPLEX # 存儲目前攝像頭中捕獲到的所有人臉的坐标/名字 # the list to save the positions and names of current faces captured pos_namelist = [] name_namelist = [] # 按下 q 鍵退出 # press 'q' to exit if kk == ord('q'): break else: # 檢測到人臉 when face detected if len(faces) != 0: # 擷取目前捕獲到的圖像的所有人臉的特征，存儲到 features_cap_arr # get the features captured and save into features_cap_arr features_cap_arr = [] for i in range(len(faces)): shape = predictor(img_rd, faces[i]) features_cap_arr.append(facerec.compute_face_descriptor(img_rd, shape)) # 周遊捕獲到的圖像中所有的人臉 # traversal all the faces in the database for k in range(len(faces)): print("##### camera person", k+1, "#####") # 讓人名跟随在矩形框的下方 # 确定人名的位置坐标 # 先預設所有人不認識，是 unknown # set the default names of faces with "unknown" name_namelist.append("unknown") # 每個捕獲人臉的名字坐标 the positions of faces captured pos_namelist.append(tuple([faces[k].left(), int(faces[k].bottom() + (faces[k].bottom() - faces[k].top())/4)])) # 對于某張人臉，周遊所有存儲的人臉特征 # for every faces detected, compare the faces in the database e_distance_list = [] for i in range(len(features_known_arr)): # 如果 person_X 資料不為空 if str(features_known_arr[i][0]) != '0.0': print("with person", str(i + 1), "the e distance: ", end='') e_distance_tmp = return_euclidean_distance(features_cap_arr[k], features_known_arr[i]) print(e_distance_tmp) e_distance_list.append(e_distance_tmp) else: # 空資料 person_X e_distance_list.append(999999999) # 找出最接近的一個人臉資料是第幾個 # Find the one with minimum e distance similar_person_num = e_distance_list.index(min(e_distance_list)) print("Minimum e distance with person", int(similar_person_num)+1) # 計算人臉識别特征與資料集特征的歐氏距離 # 距離小于0.4則标出為可識别人物 if min(e_distance_list) < 0.4: # 這裡可以修改攝像頭中标出的人名 # Here you can modify the names shown on the camera # 1、周遊檔案夾目錄 folder_name = 'E:/PersonRecognitionDlib/Person' # 最接近的人臉 sum=similar_person_num+1 key_id=1 # 從第一個人臉資料檔案夾進行對比 # 擷取檔案夾中的檔案名:LQH、YYQX、WY、WL... file_names = os.listdir(folder_name) for name in file_names: # print(name+'->'+str(key_id)) if sum ==key_id: #winsound.Beep(300,500)# 響鈴：300頻率，500持續時間 name_namelist[k] = name[0:]#人名删去第一個數字（用于視訊輸出辨別） key_id += 1 # 播放歡迎光臨音效 #playsound('D:/myworkspace/JupyterNotebook/People/music/welcome.wav') # print("May be person "+str(int(similar_person_num)+1)) # -----------篩選出人臉并儲存到visitor檔案夾------------ for i, d in enumerate(faces): x1 = d.top() if d.top() > 0 else 0 y1 = d.bottom() if d.bottom() > 0 else 0 x2 = d.left() if d.left() > 0 else 0 y2 = d.right() if d.right() > 0 else 0 face = img_rd[x1:y1,x2:y2] size = 64 face = cv2.resize(face, (size,size)) # 要存儲visitor人臉圖像檔案的路徑 path_visitors_save_dir = "E:/PersonRecognitionDlib/visitor/known" # 存儲格式：2019-06-24-14-33-40LQH.jpg now_time = time.strftime("%Y-%m-%d-%H-%M-%S", time.localtime()) save_name = str(now_time)+str(name_namelist[k])+'.jpg' # print(save_name) # 本次圖檔儲存的完整url save_path = path_visitors_save_dir+'/'+ save_name # 周遊visitor檔案夾所有檔案名 visitor_names = os.listdir(path_visitors_save_dir) visitor_name='' for name in visitor_names: # 名字切片到分鐘數：2019-06-26-11-33-00LQH.jpg visitor_name=(name[0:16]+'-00'+name[19:]) # print(visitor_name) visitor_save=(save_name[0:16]+'-00'+save_name[19:]) # print(visitor_save) # 一分鐘之内重複的人名不儲存 if visitor_save!=visitor_name: cv2.imwrite(save_path, face) print('新存儲：'+path_visitors_save_dir+'/'+str(now_time)+str(name_namelist[k])+'.jpg') else: print('重複，未儲存！') else: # 播放無法識别音效 #playsound('D:/myworkspace/JupyterNotebook/People/music/sorry.wav') print("Unknown person") # -----儲存圖檔------- # -----------篩選出人臉并儲存到visitor檔案夾------------ for i, d in enumerate(faces): x1 = d.top() if d.top() > 0 else 0 y1 = d.bottom() if d.bottom() > 0 else 0 x2 = d.left() if d.left() > 0 else 0 y2 = d.right() if d.right() > 0 else 0 face = img_rd[x1:y1,x2:y2] size = 64 face = cv2.resize(face, (size,size)) # 要存儲visitor-》unknown人臉圖像檔案的路徑 path_visitors_save_dir = "E:/PersonRecognitionDlib/visitor/unknown" # 存儲格式：2019-06-24-14-33-40unknown.jpg now_time = time.strftime("%Y-%m-%d-%H-%M-%S", time.localtime()) # print(save_name) # 本次圖檔儲存的完整url save_path = path_visitors_save_dir+'/'+ str(now_time)+'unknown.jpg' cv2.imwrite(save_path, face) print('新存儲：'+path_visitors_save_dir+'/'+str(now_time)+'unknown.jpg') # 矩形框 # draw rectangle for kk, d in enumerate(faces): # 繪制矩形框 cv2.rectangle(img_rd, tuple([d.left(), d.top()]), tuple([d.right(), d.bottom()]), (0, 255, 255), 2) print('\n') # 在人臉框下面寫人臉名字 # write names under rectangle for i in range(len(faces)): cv2.putText(img_rd, name_namelist[i], pos_namelist[i], font, 0.8, (0, 255, 255), 1, cv2.LINE_AA) print("Faces in camera now:", name_namelist, "\n") #cv2.putText(img_rd, "Press 'q': Quit", (20, 450), font, 0.8, (84, 255, 159), 1, cv2.LINE_AA) cv2.putText(img_rd, "Face Recognition", (20, 40), font, 1, (0, 0, 255), 1, cv2.LINE_AA) cv2.putText(img_rd, "Visitors: " + str(len(faces)), (20, 100), font, 1, (0, 0, 255), 1, cv2.LINE_AA) # 視窗顯示 show with opencv cv2.imshow("camera", img_rd) # 釋放攝像頭 release camera cap.release() # 删除建立的視窗 delete all the windows cv2.destroyAllWindows() Dlib模型實作人臉識别一、安裝dlib庫二、利用dlib實作人臉68個關鍵點檢測并标注三、人臉特征提取四、人臉識别參考連結 Dlib模型實作人臉識别一、安裝dlib庫二、利用dlib實作人臉68個關鍵點檢測并标注三、人臉特征提取四、人臉識别參考連結
參考連結
- python3.7添加dlib子產品
- python+OpenCv+dlib實作人臉68個關鍵點檢測并标注
- 基于dlib庫人臉特征提取【建構自己的人臉識别資料集】